Abstract

This study analyzed a log file capturing users' queries executed in the Web site of the University of Tennessee, Knoxville during March, 1997. The purpose of the study is three-fold: to understand what information needs the users of this Web site have, to investigate how successful these end-users are in searching for information, and to identify problems related to unsuccessful queries. Content analysis of each query focused on the type of information needs and the type of errors that caused a zero-hit result. Fifteen classes of information needs are identified based on content analysis of the queries; the most frequently occurred queries are searching for institutional unit and searching for academic information (counting for 40.0% ofthe total queries searched). The unsuccessful queries is more than 33.5% measured by zero-hit outcomes. Two types of errors that caused zero-hit are identified: syntactic and semantic errors. Syntactic errors occurred more often than semantic errors (53.6% vs. 46.4%). The fmdings suggest that end-users of Web resources need guidance and help in performing searches. Syntactic errors may be corrected by the search engine automatically, while semantic errors need a better information representation scheme from the Web site.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call